Increasing the Computational Feasibility of Urban Air Quality Models that Employ Complex Chemical Mechanisms

Abstract

A procedure for increasing the computational feasibility of detailed photochemical air pollution models is described. The procedure involves the development of an empirical or physically-based model that can replace the large set of stiff ordinary differential equations (ODE) that describe the chemical kinetics. The costly, time-consuming solution of these ODEs can severely limit the usefulness of general-purpose urban air quality models. Unlike conventional means of approximation such as the Empirical Kinetic Modeling Approach (EKMA), the procedures described herein can be applied in a straightforward manner to any chemical mechanism, and the resulting chemical submodel can be employed in any algorithm describing the other physical phenomena of importance. The approximate model can also be employed to conveniently determine the effect of changing precursor concentrations on ozone or other pollutant levels.

As an illustrative example, the procedures are applied to the Falls-Seinf eld-McRae mechanism1 and included In a simple dynamic box model describing transport processes. For this system, the ozone predictions of the approximate model are typically within 10-20 percent of the predictions of the full model.